US6495821B1

US6495821B1 - Method and apparatus for determining a position of a movable barrier

Info

Publication number: US6495821B1
Application number: US09/505,321
Authority: US
Inventors: Dane Smith; Chris M. Valente; Martin Rathgeber; Mark D. Siegler
Original assignee: Chamberlain Group Inc
Current assignee: Chamberlain Group Inc
Priority date: 1999-02-17
Filing date: 2000-02-16
Publication date: 2002-12-17
Anticipated expiration: 2019-02-17
Also published as: AU774964B2; EP1161795B1; AU3596800A; WO2000049716A1; AU774964C; DE60010240T2; EP1161796A4; EP1161795A1; US6376832B1; AR024856A1; NZ513727A; CA2362958C; CA2362958A1; WO2000049716A9; DE60010240D1; EP1161796A1; AU2994600A; AU778987B2; NZ513730A; CA2362708A1

Abstract

A movable barrier operator includes an absolute position detector which provides a unique value for each position of the barrier along its path of travel. The absolute position detector employs multiple binary serial streams and one multiple clock stream. After the first five cycles of the clock stream, the binary streams can be decoded by a processor to produce an absolute position. Every clock edge produces a new absolute position along the path of travel.

Description

This is a continuation-in-part, of prior application Ser. No. 09/251,307, filed Feb. 17, 1999, now abandoned, which is hereby incorporated herein by reference in its entirety.

BACKGROUND OF THE INVENTION

1. Field of the Invention

The invention relates to operators for movable barriers, such as rolling shutters, retractable awnings, gates, garage doors, overhead doors and the like, and more particularly to operators which can determine the absolute position of the barrier at all times, including after a power outage and subsequent manual relocation of the barrier.

2. Description of Related Art

One of the problems which must be addressed in designing and engineering operators for movable barriers is the provision of barrier position detection. Most electronic positioning systems used in barrier operators keep track of the barrier's position by incrementing a position counter during one direction of travel and decrementing the position counter during the opposite direction of travel. This can cause errors if there are missed pulses or extraneous pulses during travel (such as from slippage of the barrier or motor).

Some barrier position detection systems employ a pass point. The pass point corresponds to a fixed location on the barrier, so that whenever the barrier moves past the pass point, the position detector is normalized or calibrated. By normalizing or zeroing out the position detector (or counter), the effects of missed pulses or slippage are eliminated. Some systems employ multiple pass points which provide further error removal capability. The pass point is a good solution in most situations, such as for garage door operators, which seldom move manually.

A more significant problem can occur in motorized awnings or rolling shutters. The rolling shutter assembly is frequently installed in a housing which is built into a wall. If power goes out on a rolling shutter system, the user will frequently move the rolling shutter manually to either open or close it. The power is off, but the gears of the positioning system move without power applied to assure manual override of an electric system for the purpose of power failures. Some users may also decide for convenience to move the shutter manually. When power returns, if the rolling shutter has been manually moved past all pass points, the operator, not encountering the pass point reference, may cause the rolling shutter to continue to move completely into the housing necessitating removal of the shutter from the housing. Removal of the rolling shutter from the housing frequently means removing a portion of an interior wall.

There is a need for a movable barrier operator with a position indicating system that provides the absolute position of the barrier, even after power outages or after the barrier has been moved manually. There is a need for a movable barrier operator which can unambiguously determine the position of the barrier after power is applied. There is a need for a movable barrier operator which can unambiguously determine the position of the barrier regardless of direction of travel.

SUMMARY OF THE INVENTION

A barrier operator position detector includes a first rotary member which is encoded to generate a first N bit subcode selected from N sequential bits of a M bit code word. The first N bit subcode has the property that every selected subcode of N sequential bits of the M bit code word has a unique value. N is greater than 1 and preferably 5. M is greater than N and preferably 32. A second rotary member is encoded to generate a second N bit subcode selected from N sequential bits of a M−1 bit code word, the second N bit subcode also has the property that every selected subcode of N sequential bits of the M−1 bit code word has a unique value (preferably M−1 is 31). A controller, responsive to the first subcode and the second subcode, generates a 2N bit multibit (or two N-bit subcodes) code. The 2N bit multibit subcode is representative of a unique position output, which can be decoded into a unique position of the barrier along its travel.

A movable barrier operator according to the invention includes an absolute position detector which provides a unique value for each position of the barrier along its path of travel. The absolute position detector employs two binary serial streams and one clock stream. After the first five cycles of the clock stream, the binary serial streams can be decoded by a microprocessor or other processor to produce an absolute position indication. Every clock edge produces a new absolute position value along the path of travel.

The absolute position detector employs three wheels; two data wheels and a clock wheel driven by a pinon. Each wheel rotates near a wheel state detector which produces digital signals comprising bit streams. Preferably an infrared emitter-sensor pair is used as the wheel state detector. However, any electromechanical system which produces a digital signal comprising bit streams, such as Hall sensors, laser discs, and so on, may be used. For convenience, the absolute position detector of the invention will be described in detail with reference only to the infrared emitter-sensor embodiment.

In the preferred embodiment, two of the wheels are data wheels or gears and have teeth distributed around their outer portions. One wheel has 32 teeth, the other wheel has 31 teeth. Each tooth of each data wheel has a corresponding data bit formed in the wheel before the tooth. Each data bit represents a single binary data bit. A space formed below a tooth represents a digital low; a solid area formed below the tooth represents a digital high. The 32 teeth wheel has a 32 bit binary stream formed in it. The stream is uniquely defined so that any consecutive 5 bits in the stream are different from any other consecutive 5 bits in the stream, including the rollover stream. The 31 teeth wheel is similarly defined, except the 31 teeth wheel has the same bit binary stream as the 32 bit wheel, with one bit missing.

Since the 32 teeth wheel and the 31 teeth wheel have different numbers of teeth and are driven by the same pinion, they rotate at different speeds. The pinion is driven externally by a gearing system that is driven by the motor. The motor can rotate clockwise or counterclockwise, so the pinion can also turn in both directions. Since the motor is bi-directional, an attached load comprising a barrier such as a door, awning, shutter or gate can move in either of two opposite directions. Preferably direction of travel information is obtained by storing the commanded direction of travel (e.g., the user commands the door to open by pushing the open button or to close by pushing a close button).

The movable barrier is operated through linear linkage of the load to the motor such as a trolley, or through rotational linkage to the motor, where the load is wound around the entire operator unit, such as in a rolling shutter. Since the two data wheels rotate at different speeds (because of the different number of teeth), the two binary streams have different repeat rates. This means that a given 5 bit stream from the 32 teeth wheel will not combine with the corresponding 5 bit stream from the 31 teeth wheel until 31 more revolutions of the 32 teeth wheel, or vice versa. In other words, a total of 31×32=992 unique two word values are possible without a rollover or repeated position concern. 992 unique positions is large enough to provide absolute position along a part of a movable barrier in most situations. An extra data wheel may be added for more positions (e.g., 32×31×30=29,760 positions). This mechanical linkage also means if the unit is moved manually, the 31 bit wheel and the 32 bit wheel will move, storing or representing for later reading by the controller, the position of the awning, door or shutter.

The third wheel is a clock wheel and is used to provide a clock signal for the position detecting system to enable proper sampling of the data wheel bit streams. The clock wheel includes 32 equally spaced openings. The clock wheel provides a digital low pulse signal when the center of a data bit on the 32 teeth wheel lines up with the center of a data bit on the 31 teeth wheel and when these centers are in line with the IR sensors. The clock signal is provided to the microprocessor which uses the clock signal as an interrupt to sample binary data from emitter-receiver pair associated with each data wheel. After the first 5 clock cycles, each data wheel has output a 5 digit binary stream, which when combined, gives 2 five digit binary numbers. This 5 digit binary number pair is decoded by the microprocessor which calculates an absolute position. Thereafter, every clock cycle triggers the sampling of a new binary digit from each wheel, the stored 5 bit binary number pair is updated, and a new absolute position of the barrier is determined.

The movable barrier operator according to the invention with absolute position detector (or encoder system) provides many advantages. It provides the absolute position of the barrier for every pulse edge of the clock signal from positioning gears turned by motor's gearing system. Shortly after power is applied to the motor, the clock wheel would have produced 5 pulses. After 5 pulses, the encoder system determines the absolute position of the barrier. The encoder system can provide direction of travel after six pulses of the clock wheel. The encoder system discriminates false or unwanted pulses to prevent false positioning.

An absolute position is always provided shortly after power is applied (after 5 pulses and the first 5 digit binary pair is obtained), regardless of the stored value of the last position and regardless of where the barrier may have been moved manually. The absolute position detector also provides an opportunity for the system to do a validity check for every newly calculated position (e.g., by checking the absolute position between successive data streams, the direction of travel can be ascertained). It should also be noted that no presetting of the wheels prior to installation/operation is required.

Additional advantages and features of the invention may be appreciated from a perusal of the specification, including claims in light of the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a perspective view of a garage door operating system in accordance with an embodiment of the invention;

FIG. 2 is a perspective view of a rolling shutter operating system in accordance with an alternative embodiment of the invention;

FIG. 3 is a perspective view of the tubular motor assembly of FIG. 2;

FIGS. 4 and 5 are two exploded perspective views of the location of the absolute position detector assembly shown in FIG. 3;

FIG. 6 is an enlarged perspective view of the absolute position detector assembly of FIG. 4;

FIG. 7 is a graph of the 32 bit data streams produced in each of the 31 bit wheel and 32 bit wheel;

FIG. 8 is an example calculation of position using the 31 bit wheel and the 32 bit wheel;

FIG. 9 is a flow chart of the routine run by the controller to sample the 5 bit data streams;

FIG. 10 is a flow chart of the RPM routine used by the controller to sample the 5 bit data streams;

FIGS. 11A-C are schematic diagrams of the electronics controlling the rolling shutter head unit of FIG. 2.

FIGS. 12A-C is a flow chart of the routine used to generate the unique binary stream;

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

Referring now to the drawings, and especially to FIG. 1, a movable barrier operator embodying the present invention is generally shown therein and identified by reference numeral 10. The movable barrier operator 10 is employed for controlling the opening and closing of a conventional overhead garage door 12 of a garage 13. The garage door 12 is mounted on guide rails 14 for movement between the closed position illustrated in FIG. 1 and an open or raised position. The garage 13 includes a ceiling 16 and a wall 18 defining an opening blocked by garage door 12. As shown, guide rails 14 are mounted to wall 18 and ceiling 16 of the garage 13 in a conventional manner.

A power drive unit or head, generally indicated at 20, is mounted to the ceiling 16 in a conventional manner. An integrated drive rail 22 extends between the power drive unit 20 and the garage wall 18. As can be seen in FIG. 1, one end of integrated drive rail 22 is mounted to a portion of the garage wall 18 located above the garage door 12. An operator arm 26 is connected at one end to the garage door 12 and at the other end to a trolley 28 mounted for movement back and forth, along the integrated drive rail 22. As will be seen herein, a motor in the power drive unit 20 propels the trolley 28 in a desired manner to raise and lower garage door 12 via the coupling of the trolley 28 and the operator arm 26 to the garage door 12.

A push button control unit 32, which includes an electronic controller and a keypad, is coupled by electrical conductors 34 to the power drive unit 20 and sends signals to the power drive unit, controlling operation of the drive motor therein. Preferably, the power drive unit 20 also includes a conventional radio receiver (not shown) for receiving radio signals from a remote control transmitter 38. An optional auxiliary power drive unit 40 is shown coupled to one end of integrated drive rail 22, being mounted on wall 18, atop door 12. If desired, operational flexibility of the integrated drive rail assembly may allow relocation of the main drive unit to a point adjacent the door.

Referring now to FIG. 2, a barrier operator system employing an absolute position detector is employed for controlling the opening and closing of a conventional rolling shutter 112. The rolling shutter is mounted on guide rails 114 for movement between the closed position illustrated in FIG. 2 and an open or raised position. The wall 118 defines an opening blocked or covered by rolling shutter 112. As shown, guide rails 114 are mounted to wall 118 in a conventional manner.

A power drive unit or head, generally indicated at 120, is mounted to the top of frame 110 in a conventional manner. Although the head unit is shown as being mounted on the exterior, as noted above, in many applications, the head unit is built into the wall so the user sees only the shutters. In the two views shown in FIG. 2, the head unit 120 is shown mounted on opposite sides of the top of frame 110. As will be seen herein, a motor in head unit 120 propels a sleeve or tube 142 to raise and lower rolling shutter 112 via the coupling of sleeve 142 to rolling shutter 112.

Control for head unit 120 may be as described above for garage door operator 20, e.g., using a push button control or a keypad mounted at another location on a wall. Additionally, head unit may also include a conventional radio receiver (not shown) for receiving radio signals from a remote control transmitter. If desired, the head unit 120 may be mounted on either side of the frame 110.

As shown in FIGS. 3, 4 and 5, head unit 120 includes a tubular housing 138 and

end section

122 and 134. Within the housing 138 is the motor 130 which includes an output shaft 131 coupled at one end to end section 134 and at the other end to driving gear assembly 132. The output from gear assembly 132 is provided to output ring 140, which is fixedly attached to outer sleeve 142. Rolling shutters are attached to outer sleeve 142, so that when motor 130 runs, outer sleeve 142 rotates, causing rolling shutters to open or close (depending on the direction of rotation of motor 130).

Outer sleeve

142 is also fixedly attached to ring 136. Ring 136 drives absolute position detector assembly 124. Position detector assembly 124 is coupled to control board 144. Control board 144 contains the electronics for starting and controlling motor 130 (see FIGS. 11A-C). Capacitor 126 is used to start motor 130 (described below). A brake 128 is provided to slow motor 130 when the rolling shutters are approaching a limit position. 126 is used to start motor 130 (described below). A brake 128 is provided to slow motor 130 when the rolling shutters are approaching a limit position.

Referring to FIGS. 6 and 7, absolute position detector assembly 124 includes a clock wheel 206, which is attached to axle 212 for rotation therewith. Axle 212 rests in supports 210, and freely rotates therein, which are attached to board 144 by legs 240. Clock wheel 206 includes 32 equally spaced openings 230. The clock wheel 206 provides a digital low pulse signal when the center of a data bit on the 32 teeth wheel 202 lines up with the center of a data bit on the 31 teeth wheel 204 and when these centers are in line with the IR sensors—through an opening 230 (not shown). The clock signal is provided to the microprocessor which uses the clock signal as an interrupt to sample binary data from each data wheel. 32 bit wheel 202 is attached to axle 212 for rotation therewith. Each complete rotation of the 32 bit wheel 202 corresponds to one complete rotation of clock wheel 206. 32 bit wheel 202 includes 32 teeth or gears 220, which are driven by pinion 252 (see FIG. 4) which is driven by ring 136. 31 bit wheel 204 includes 31 teeth or gears 222 which are also driven by pinion 252. 31 bit wheel 204 freely spins about axle 202. One turn of the 32 bit wheel 202 corresponds to 32/31 turns of the 31 bit wheel 204.

A unique bit stream pattern is formed in each of 32

bit wheel

202 and 31 bit wheel 204. Beneath the teeth 220 are solid areas 224 and spaces 226. A space under a tooth 220 corresponds to a 0; a solid area 224 correspond to a 1. The exact pattern is shown in FIG. 7. The first row of pulses are the 32 pulses generated by the clock wheel 206. One complete revolution of the clock wheel generates 32 low pulses, representing sample time. The 31 bit wheel has solid and space areas which correspond to a 31 bit data stream: 1111000001110100010010101100110 as shown in the second row of FIG. 7. For every one complete revolution of the clock wheel, the 31 bit wheel produces the unique 31 bit data stream plus one rollover bit. The 32 bit wheel 202 generates the data stream: 11111000001110100010010101100110, which is the same pattern as the 31 bit data stream with the addition of an extra 1 at the beginning of the stream. This data stream is constant for every revolution of the clock wheel.

In the 32 bit stream, no five consecutive bits are repeated anywhere else in the stream. This is also true for the 31 bit data stream. When the unit is powered for movement, five consecutive (or sequential) bits are sampled from each wheel. The decimal value is calculated for each 5 bit number. The lookup table A (attached hereto) is used to convert the 5 bit number to a decimal number. Then a mathematical operation is performed on the two converted numbers (from the 31 bit wheel and the 32 bit wheel) to produce an absolute position.

Referring to FIG. 7, if the unit were powered up with the wheels aligned as shown in FIG. 7, the first 5 bit data stream sampled would be: 11110 for the 31 bit wheel and 11111 for the 32 bit wheel. In the next clock cycle, after rotation of 1/32 of the clock wheel a clock pulse is generated, the 31 bit wheel produces 11100 and the 32 bit wheel produces 11110. Continuing for 32 1/32 steps, or 32 5 bit frames, each sequential or consecutive 5 bit data stream produced by each wheel is unique.

An example calculation is shown in FIG. 8. A 5 bit data stream is sampled from each of the 31 bit wheel and the 32 bit wheel. In this example, the 31 bit wheel produces the 5 bit data stream: 01000. The 32 bit wheel produces the 5 bit data stream 10101. These numbers convert to 08 (Lookup1) 21 (Lookup2), respectively, using the lookup table A. 12-20=−8. If the result is negative, add 31 (Same as modulo 31 arithmetic). Apply the mathematical formula: (Result×32)+Lookup2=Absolute position. This gives an absolute position of 756 out of 992 possible positions along the path of travel.

The calculation of absolute position is performed in two interrupt routines by the controller. The first interrupt routine samples the clock and data wheels and generates the next bit to be used in the sliding window or sliding 5 bit data stream. When the clock wheel generates a digital low pulse, the controller executes the absolute position routine, shown in FIG. 9. Referring to FIG. 9, at step 300, the routine checks if the IR sensor and detector are operational. If the IR sensor and detector are not operational, the controller leaves the routine at step 318. If the IR sensor and detector are operational, the routine checks if the motor is on at step 302. If not, the routine exits at step 318. If the motor is on, the routine checks at step 304 if the clock pulse is going low, indicating the beginning of a clock pulse. If not, the routine exits at step 318.

If the clock pulse is going low, the routine sets the state of the 31 bit wheel (WHEEL_31_STATE) register and the state of the 32 bit wheel state (WHEEL_32_STATE) register low in step 306. These registers store the value of the next detected data bit. At step 308, the routine checks if the 31 bit wheel stream is high. If yes, it sets the 31 bit state register to high in step 310. If not, it continues to block 312 where it checks if the 32 bit wheel stream is high. If yes, it sets the 32 bit wheel state register to high at step 314. If not, it calls the RPM routine, then leaves the routine at step 318. The RPM routine takes the current bit and uses it to create the next 5 bit data stream for use in calculating the absolute position of the shutter.

Once the 5 bit streams are computed and stored, the controller computes the absolute position as described above and uses that information to keep track of where the door or shutter is at each clock cycle and as a validity check for direction of movement. It should be noted that if the awning, door or shutter is moved manually, movement of the door or shutter will drive the pinions moving the clock wheel and 31 bit wheel and 32 bit wheel, so door/shutter position is always mechanically recorded in the absolute position detector assembly, ready for reading when the unit is powered on.

After the current bit from each wheel is stored in the appropriate register, the RPM routine is called. Referring to FIG. 10, at step 340, the routine checks for the direction of travel. This information is typically provided by the user input when the user selects the up button or down button. As noted above, this information can be verified and/or changed if the absolute position information does not check out between successive clock pulses.

If the shutter is moving up the routine branches to step 344. If the shutter is moving down, the routine branches to step 342. Each

step

342 and 344 forms the appropriate sliding window (determines the consecutive 5 bits to be used in calculating the shutter position). In step 344 the routine shifts the MASK 31 bits left. The MASK_31 bit mask is a window of all 31 bits the 31 bit wheel. Then the least significant bit of the MASK_31 is logically OR'd with the 31 bit wheel state register. Only the first 5 bits of the MASK_31 mask (which contains the entire 31 bit data stream represented on the 31 bit wheel) are masked. Then the MASK_32 bit mask (which contains the entire 32 bit data stream represented on the 32 bit wheel) is shifted left one bit and the least significant bit of the MASK_32 bit mask is logically OR'd with the value in the 32 bit wheel state register. Only the first 5 bits are masked. This gives two shifted 5 bit data streams, one each from the 31 bit wheel and the 32 bit wheel, which are used to determine the position of the shutter for that clock cycle.

In step 342 the routine shifts the MASK_31 bits right. Then the 5th least significant bit of the MASK_31 is logically OR'd with the WHEEL_31_STATE register. Then only the first five least significant bits of the MASK_31 are masked. The MASK_32 mask is shifted one bit right. Then the MASK_32 mask is logically OR'd with the WHEEL_32_STATE register.

In step 346 the routine uses a ROM lookup table (see Table A) to get a conversion for the numbers in MASK_31 and MASK_32. These digital numbers are stored in the variables MASK_31_VALUE and MASK_32_VALUE. In step 348, the difference between MASK_31_VALUE and MASK_32_VALUE is calculated and the remainder from modulo 31 arithmetic calculated. This result is called the DIFFERENCE. In step 350 the DIFFERENCE is multiplied by 32. Then MASK_32_VALUE is added to the product. This number is the absolute position and is stored in the POS_CNTR. At step 354 the routine ends.

The controller uses the POS_CNTR value in controlling the operation of the shutter in its other routines, which are not described.

A schematic of the control circuit located on control board 142 is shown in FIG. 11. Controller 500 operates the various software routines which operate the rolling shutter operator 120. Controller 500 may be a Z86733 microprocessor. In this particular embodiment, the rolling shutter is controlled only by a wall-mounted or unit-mounted switch coupled via connector J2. Connector J2 has inputs for up switched hot and down switched hot. In a rolling shutter, the motor moves only when the user presses the power direction switch connected to connector J2 and the Triac Q1 is activated by the microcontroller. Pressing the up or down switch applies power to the board via connector J2 and provides various motor phase and direction information to the controller 500. When the controller 500 permits travel, Triac Q1 enables the motor's neutral path. The motor winding, which is then powered, will conduct current.

However, the control circuit can be modified to include a receiver so that the rolling shutter can be commanded from a remote transmitter (as described above). Power supply circuit 190 converts AC line power from connector J2 into plus 5 volts to drive the logic circuits and plus 16 volts for a voltage supply to the phototransistors Q4, Q5, Q6.

Upon receipt of a rolling shutter movement command signal through J2, the motor is activated. Feedback information from the motor and AC power is provided from J1 and applied to U3:A, U3:B, U3:C and U3:D. The outputs from U3:B and U3:D provide up and down phase information to pins P26 and P25 respectively. The outputs from U3:A and U3:C provide up and down direction to pins P21 and P20, respectively.

Crystal CR1 provides an internal clock signal for the microprocessor 500. EEPROM 200 stores the information such as limit flags, force flags, learn mode flags, etc. The IR signal break from clock wheel 206 drives Q5 which provides an input to signal P31. Wheel 31 drives Q4 which provides an input signal to P30. Wheel 32 drives Q3 which provides an input signal to P33.

FIGS. 12A-C are a flow chart illustrating the algorithm for generating a unique binary stream, or an array of decimal values for any given number to the power of two, (i.e., 32, 64, 128, etc.). More particularly, this algorithm generates the binary stream that is listed in Table A and can be used to design the 32 bit wheel located between the IR emitter/receiver pair. As FIGS. 12A-C indicate, the algorithm for generating the unique non-repeating data stream or binary serial stream, is a very simple one. Although the algorithm depicted in FIGS. 12A-C can be used to design any bit wheel that is to the power of two, (e.g., 2⁵, 2⁶, 2⁷, etc.), the example provided assumes a 32 bit wheel is used. The algorithm begins at an initial state, step 400, where the repeat flag is set to zero (REAPEAT_FLAG=0), the branch flag is set to zero (BRANCH_FLAG=0), the initial value is set to zero (i=0), array A is set equal to the highest value allowed (A(i)=31), and array B is set to zero (B(i)=0). At step 300, Array A sub-zero is equal to 31 (Array A(0)=31), which is the highest value allowed in a 32 bit system. Array B holds the branching flag for branching routine and will be discussed in further detail at step 442. In short, the branching flag of array B indicates whether a number has been used previously in algorithm A, or if it has not already been used in array A which would mean it is a number that could be used in the future. Moreover, the decimal values listed in column two of Table A could have been listed as their binary equivalent as the algorithm works the same for either decimal values or their binary equivalents.

At step 402, the number in array A is doubled and masked (A(i+1)=2*(A(i) Mask 0011111B). Since our initial value is zero, this step sets A sub-one equal to decimal value thirty (which is the equivalent of sixty-two masked). Decision step 404 is then entered in which the algorithm asks if the branch flag is equal to one (BRANCH_FLAG=1?). Since this is the initial run of the algorithm (i=0), the branch flag was set to zero in step 400, so the answer to this decision block is no. However had this been any later run (i=1, 2, etc.) the branch flag would not have been set equal to zero in step 400. The branch flag is set in the check branch subroutine (CHECK_BRANCH) discussed below. In short, if the branch flag is equal to zero, there is no possibility the current value of array A has been used before (or there is no possibility the current number is recursive). If the branch flag is equal to one, there is a possibility that the number is a repeat of an earlier number. The significance in this determination is that the barrier operator position detector of the invention generates a unique non-repeating binary serial stream.

If the branch flag in step 404 is not equal to one, control shifts to step 406 and the algorithm enters the check repeat subroutine (CALL CHECK_REPEAT). During the check repeat subroutine step 408 (FIG. 12B), a number n is assigned the value of i. This is done because the value of n will change during the subroutine, and we do not want to alter the value of i. A dummy variable is named COMPARE and is set equal to A sub-n plus one (COMPARE=A(n+1). Since i is equal to zero, n is equal to zero and COMPARE equals A sub-one. Control is shifted to step 410 and the algorithm asks if COMPARE is equal to A(n) (COMPARE=A(n)?). Basically this step checks to see if the variable COMPARE is a repeat of any of the previous numbers in the array. If COMPARE is equal to A sub-n, (meaning it is a repeat of an earlier number in the array), control shifts to step 412, where the repeat flag is set equal to one (REPEAT_FLAG=1) indicating that this number has already been used. After setting the repeat flag equal to one in step 412, control is shifted to step 414, which shifts control back to main routine step 406. If COMPARE is not equal to A sub-n, (meaning it is not a repeat of the previous number in the array), control is shifted to decision step 416 in which the algorithm asks if n is equal to zero (n=0?). If n is not equal to zero, there are other numbers in the array to compare COMPARE to. Therefore control is shifted to step 418 where n is decremented (dec n) and control is sent back to step 410. Once back at step 410, the algorithm again asks if COMPARE is equal to a number previously used in the array. This process is continued until COMPARE has been checked against every number in the array. If COMPARE is a repeat of any of the previous numbers used in the array, control shifts from step 410 to step 412. However, if COMPARE is not equal to any number previously used, and n is equal to zero (indicating all the numbers previously used in the array have been checked), control shifts from step 416 to step 420. Step 420 sets the repeat flag equal to 0 (REPEAT_FLAG=0), transfers control to return step 414, which transfers control back to step 406 in the main routine.

Once control returns from the check repeat subroutine to step 406, the algorithm checks to see if the repeat flag is equal to one (REPEAT_FLAG=1?). If the repeat flag is not equal to one, the present even number has not been repeated previously in the array (by default the system starts with an even number). Therefore, control shifts to step 422 in which the algorithm increments the even number to make it odd so that a further check can be made to see if the odd number is repetitive of a number previously used in the array. Specifically, A sub-i plus one is incremented to test the odd number (A(i+1)=A(i+1)+1). After step 422 has created the odd number, control is shifted to step 424 in which the algorithm again calls the check repeat subroutine (CALL CHECK_REPEAT). The odd number is run through the same steps 408-414 of the check repeat subroutine as the even number was and control is eventually returned back to decision step 424. Once the subroutine is exited, the algorithm asks if the repeat flag is equal to one (REPEAT_FLAG=1?). If the repeat flag is equal to one, (meaning the odd number is repetitive of a number previously used in the algorithm), control is shifted to step 426. Step 426 sets the B array equal to zero, so that the algorithm knows that this is not a possible branch number or flag (meaning the number has already been used, so we do not want to indicate it as a number that can be branched back to for future numbers). Once the branch flag has been set equal to zero, control is shifted to step 428 so the A sub-i plus one can be converted back to the original even number that existed prior to step 422. Specifically, A sub-i plus one is set equal to A sub-i plus one minus one (A(i+1)=A(i+1)−1). Once the original even number is restored, control is transferred to step 430.

If in step 424, the repeat flag is not equal to one, control will be transferred to step 432 instead of step 426. In step 432, the check repeat subroutine has discovered that the odd number is not repetitive of any of the numbers already used in the array and therefore could be a number used later on in the array. Step 432 indicates this by setting the B array equal to one (B(i+1)=1). Once this has been done, control is shifted to step 428 which, as mentioned earlier, converts the odd number back into the original even number that existed prior to step 422. Specifically, A sub-i plus one is set equal to A sub-i plus one minus one (A(i+1)=A(i+1)−1). Once the original number is restored, control is transferred to step 430.

If the branch flag of step 404 is equal to one or step 406 detects that the repeat flag is equal to one, the present even number has been repeated previously in the array. Control is then shifted to step 434 so that array B, branch flag and repeat flag can be set equal to zero. Once this has been done, control shifts to step 436 in which the algorithm increments the even number to make it odd (similar to step 422) so that a further check can be made to see if the odd number is repetitive of a number previously used in the array. Specifically, A sub-i plus one is set equal A sub-i plus one plus one in order to test the odd number (A(i+1)=A(i+1)+1). After step 436 has created the odd number, control is shifted to step 438 in which the algorithm again calls the check repeat subroutine (CALL CHECK_REPEAT). The odd number is run through the same steps 408-414 of the check repeat subroutine as the even number was and control is eventually returned back to decision step 438. Once the subroutine is exited, the algorithm asks if the repeat flag is equal to one (REPEAT_FLAG=1?). If the repeat flag is equal to one, (meaning the odd number is repetitive of a number previously used in the algorithm), control is shifted to step 440 and the check branch subroutine is entered (CALL CHECK_BRANCH).

During the check branch subroutine step 442, a variable x is assigned the value of i (x=i). This is done so that the value of x can be changed without altering the value of i. Once x is established, control switches to decision step 444 in which the algorithm asks if the value of B sub-x is equal to 1 (B(x)=1?). If B sub-x does not equal one, control is transferred to 446 and the value of x is decremented (dec x). Once the value of x has been decremented by one, control shifts from step 446 back to step 444 and the algorithm again asks if the value of B sub-x is equal to 1 (B(x)=1?). This process is continued until B sub-x is equal to one, (meaning until a number is found that has not been used previously in the array). Once this number has been found, control transfers from step 444 to step 448, i is set equal to x minus one (i=x−1) and the branch flag is set equal to one (BRANCH_FLAG=1). After step 448 is completed, control transfers to step 450 and the algorithm returns from the check branch subroutine to step 402 of the main routine. If the repeat flag in step 438 is not determined to be equal to one control is transferred to step 430 so that the value of i can be incremented (Inc i). The incrementing of i is what causes the system to default to an even number.

After i has been incremented in step 403, control is transferred to step 452 to determine whether i is equal to 31 (i=31). This question is asked in order to find out if we are at the last number in the array. Since we have been discussing a 2 to the 5th power array, or a 32 bit array, the largest possible value is 31. However, if a 64 bit array was used, step 452 would ask if i is equal to 63, etc. If i is not equal to the last number in the array, control transfers from step 452 to step 402 and the process begins again. If i is equal to the last number in the array, control transfers from step 452 to step 454 and the array is printed.

The position is detected by looking at a five bit window during the revolution of the 32 bit wheel. The five bits window never repeats itself so that an accurate position can be detected. This is a result of the algorithm disclosed on FIGS. 12A-C generating an array of decimal numbers for any given number of teeth on a wheel that is a power of two. The algorithm aids in the design of the wheel that rotates between the IR emitter/receiver pair. Do to the size constraints posed by a tubular motor housing, a smaller wheel is desirable. Therefore a 32 bit array is used. However, in a garage door, a larger array is preferred, so a 64 bit array is used. The only difference being the size of the wheel and the size of the window used, (e.g., for 32 bit array a 5 bit window is needed, for a 64 bit array a 6 bit window is needed, etc.). Use of additional IR emitter/receiver pairs will assist in determining the absolute position more rapidly.

Although the above discussion and related drawings discuss use of the algorithm for generating an array of decimal values for any given number to the power of two, it is believed the same can apply to numbers to the power of one as well.

Table A attached hereto is the lookup table described above.

Exhibit A (pages A1-A21) attached hereto include a source listing of a series of routines used to operate a movable barrier operator in accordance with the present invention.

As will be appreciated from studying the description and appended drawings, the present invention may be directed to operator systems for movable barriers of many types, such as fences, gates, shutters, awnings, garage doors, overhead doors and the like.

While there have been illustrated and described particular embodiments of the invention, it will be appreciated that numerous changes and modifications will occur to those skilled in the art, and it is intended in the appended claims to cover all those changes and modifications which fall within the true spirit and scope of the invention.

Claims

What is claimed is:

1. A barrier operator position detector, comprising:

a circuit for generating a unique non-repeating binary serial stream in response to movement of a barrier and windowing said serial stream to identify a plurality of consecutive elements of the binary serial stream that identify the barrier's position, wherein at least some consecutive barrier positions have corresponding serial stream windows that partially overlap with respect to the plurality of consecutive elements of the binary serial stream.

2. A barrier operator position detector according to claim 1, wherein said circuit for generating a unique non-repeating binary serial stream comprises:

a device encoded to generate a subcode having a unique value; and

a controller responsive to said subcode and further capable of generating a subcode representative of a unique position output which is reflective of a position of the barrier in its travel.

3. A barrier operator position detector according to claim 2, wherein said encoded device further comprises:

a first rotary member encoded to generate a first five bit subcode selected from five sequential bits of a 32 bit code word, the first five bit subcode having the property that every selected subcode of five sequential bits of the 32 bit code word has a unique value; and

a second rotary member encoded to generate a second five bit subcode selected from five sequential bits of a 31 bit code word, the second five bit subcode having the property that every selected subcode of five sequential bits of the 31 bit code word has a unique value.

4. A barrier operator position detector according to claim 3, wherein said controller is responsive to the first subcode and the second subcode, for generating a ten bit multibit subcode, wherein the ten bit multibit subcode is representative of a unique position output, the position output being reflective of a position of the barrier in its travel.

5. A barrier operator position detector according to claim 4, further comprising a clock member for generating a timing signal, wherein the controller, responsive to the timing signal, samples the first subcode and the second subcode.

6. A barrier operator position detector according to claim 5, wherein the clock member comprises a light emitter, a light receiver and a rotary interrupter mechanism disposed there between for selectively interrupting transmission of light from the light emitter to the light receiver, wherein the timing signal is generated upon interruption of the light transmission.

7. A barrier operator position detector according to claim 6, wherein after five cycles of the clock, the binary serial streams can be decoded by the controller to produce an absolute position indication.

8. A barrier operator position detector according to claim 6, wherein the first rotary member further comprises 32 geared teeth and wherein the second rotary member further comprises 31 geared teeth, the gears being adapted for being drivingly connected to a shaft for rotation.

9. A barrier operator position detector according to claim 5, wherein the clock member comprises two or more light emitters, two or more light receivers and a rotary interrupter mechanism disposed there between for selectively interrupting transmission of light from the light emitters to the light receivers, wherein the timing signal is generated upon interruption of the light transmission.

10. A barrier operator position detector according to claim 9, wherein after one cycle of the clock, the binary serial streams can be decoded by the controller to produce an absolute position indication.

11. A method of detecting a barrier's position, comprising:

generating a unique non-repeating binary serial stream in response to movement of the barrier; and

windowing said serial stream to identify a plurality of consecutive elements of the binary serial stream that identify the barrier's position, wherein windowing said serial stream with respect to at least some consecutive barrier positions yields corresponding serial stream windows that partially overlap with respect to the plurality of consecutive elements of the binary serial stream.